Scholarlink

<p>Remote sensing change detection (RSCD) requires effective modeling of long-range spatial dependencies and large receptive fields to accurately capture subtle land-cover variations between bi-temporal remote sensing images. However, existing RSCD methods often suffer from limited receptive fields or high computational complexity, constraining their ability to capture large-scale contextual dependencies. To address these challenges, this paper proposes a&#xa0;Large Kernel State Space Model for Change Detection (LKMamba-CD). Within LKMamba-CD, the Large Kernel SS2D Module (LKS) is designed to extract large-scale and long-range feature correlations. The Bi-temporal Change Information Fusion Module (BCIF) is utilized to fuse bi-temporal change information. Meanwhile, the Large Receptive Field Feature Aggregation Module (LRFA) is utilized to extract large-scale land cover feature correlation. Within BCIF and LRFA, LKS serves as the core component, leveraging re-parameterized large kernel convolution and the Mamba architecture to capture long-range dependencies with expansive receptive fields efficiently. Experiments on S2Looking, WHU-CD, and GVLM-CD show that LKMamba achieves the highest F1 of 62.02%, 92.52%, and 87.58%, respectively. The code of LKMamba-CD at <a href="https://github.com/kion-86/lkmamba">https://github.com/kion-86/lkmamba</a>.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

LKMamba-CD: Large Kernel State Space Model for Remote Sensing Change Detection

  • 期刊论文

  • 发表日期：

    2026年7月24日

查看全文
  • Yi Yu,
  • Li Bo,
  • Maoxi Zheng,
  • Huan Wang

摘要

Remote sensing change detection (RSCD) requires effective modeling of long-range spatial dependencies and large receptive fields to accurately capture subtle land-cover variations between bi-temporal remote sensing images. However, existing RSCD methods often suffer from limited receptive fields or high computational complexity, constraining their ability to capture large-scale contextual dependencies. To address these challenges, this paper proposes a Large Kernel State Space Model for Change Detection (LKMamba-CD). Within LKMamba-CD, the Large Kernel SS2D Module (LKS) is designed to extract large-scale and long-range feature correlations. The Bi-temporal Change Information Fusion Module (BCIF) is utilized to fuse bi-temporal change information. Meanwhile, the Large Receptive Field Feature Aggregation Module (LRFA) is utilized to extract large-scale land cover feature correlation. Within BCIF and LRFA, LKS serves as the core component, leveraging re-parameterized large kernel convolution and the Mamba architecture to capture long-range dependencies with expansive receptive fields efficiently. Experiments on S2Looking, WHU-CD, and GVLM-CD show that LKMamba achieves the highest F1 of 62.02%, 92.52%, and 87.58%, respectively. The code of LKMamba-CD at https://github.com/kion-86/lkmamba.